LocalAI version:
quay.io/go-skynet/local-ai:v4.1.3-gpu-vulkan@sha256:ab337d593517231464cd50fe81fa4f84ee9bda5be0856e2591cdfd5f020a70bf
Environment, CPU architecture, OS, and Version:
Linux 7997ab623f44 6.19.10+deb13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.19.10-1~bpo13+1 (2026-04-02) x86_64 x86_64 x86_64 GNU/Linux
AMD Ryzen 7 3700X
Intel Arc Pro B70
Describe the bug
The VRAM for my Intel Arc Pro B70 is not detected. Inference works correctly.
To Reproduce
Start the container, see only device RAM and no VRAM.
Expected behavior
Correctly show used VRAM.
Logs
Apr 09 08:34:46 DEBUG GPU vendor detected via ghw vendor="intel" caller={caller.file="/build/pkg/xsysinfo/gpu.go" caller.L=159 }
Apr 09 08:34:46 DEBUG GPU vendor gpuVendor="intel" caller={caller.file="/build/pkg/system/state.go" caller.L=77 }
Apr 09 08:34:46 DEBUG failed to parse vulkaninfo output error=unexpected end of JSON input caller={caller.file="/build/pkg/xsysinfo/gpu.go" caller.L=795 }
Apr 09 08:34:46 DEBUG Total available VRAM vram=0 caller={caller.file="/build/pkg/system/state.go" caller.L=79 }
Apr 09 08:34:46 INFO Using forced capability run file capabilityRunFile="/run/localai/capability" capability="vulkan\n" env="" caller={caller.file="/build/pkg/system/capabilities.go" caller.L=98 }
Apr 09 08:34:46 INFO Starting LocalAI threads=8 modelsPath="//models" caller={caller.file="/build/core/application/startup.go" caller.L=38 }
Apr 09 08:34:46 INFO LocalAI version version="v4.1.3 (fdc9f7bf350b05942323f5f4b264480c7b67b51c)" caller={caller.file="/build/core/application/startup.go" caller.L=39 }
Apr 09 08:34:46 DEBUG CPU capabilities capabilities=[3dnowprefetch abm adx aes aperfmperf apic arat avic avx avx2 bmi1 bmi2 bpext cat_l3 cdp_l3 clflush clflushopt clwb clzero cmov cmp_legacy constant_tsc cpb cpuid cqm cqm_llc cqm_mbm_local cqm_mbm_total cqm_occup_llc cr8_legacy cx16 cx8 de decodeassists extapic extd_apicid f16c flushbyasid fma fpu fsgsbase fxsr fxsr_opt ht hw_pstate ibpb ibs irperf lahf_lm lbrv lm mba mca mce misalignsse mmx mmxext monitor movbe msr mtrr mwaitx nonstop_tsc nopl npt nrip_save nx osvw overflow_recov pae pat pausefilter pclmulqdq pdpe1gb perfctr_core perfctr_llc perfctr_nb pfthreshold pge pni popcnt pse pse36 rapl rdpid rdpru rdrand rdseed rdt_a rdtscp rep_good sep sev sev_es sha_ni skinit smap smca smep ssbd sse sse2 sse4_1 sse4_2 sse4a ssse3 stibp succor svm svm_lock syscall tce topoext tsc tsc_scale umip v_spec_ctrl v_vmsave_vmload vgif vmcb_clean vme vmmcall wbnoinvd wdt x2apic xgetbv1 xsave xsavec xsaveerptr xsaveopt xtopology] caller={caller.file="/build/core/application/startup.go" caller.L=47 }
Apr 09 08:34:46 DEBUG GPU count count=1 caller={caller.file="/build/core/application/startup.go" caller.L=52 }
Apr 09 08:34:46 DEBUG GPU gpu="card #0 @0000:0f:00.0 -> driver: 'xe' class: 'Display controller' vendor: 'Intel Corporation' product: 'unknown'" caller={caller.file="/build/core/application/startup.go" caller.L=54 }
Additional context
vulkaninfo inside container shows the correct VRAM (although usage is 0). The VRAM is currently loaded with ~24GiB.
VkPhysicalDeviceMemoryProperties:
=================================
memoryHeaps: count = 2
memoryHeaps[0]:
size = 34242297856 (0x7f9000000) (31.89 GiB)
budget = 7537164288 (0x1c1400000) (7.02 GiB)
usage = 0 (0x00000000) (0.00 B)
flags: count = 1
MEMORY_HEAP_DEVICE_LOCAL_BIT
memoryHeaps[1]:
size = 101233557504 (0x1791fd8000) (94.28 GiB)
budget = 58268319744 (0xd91100000) (54.27 GiB)
usage = 0 (0x00000000) (0.00 B)
flags:
None
memoryTypes: count = 7
memoryTypes[0]:
heapIndex = 0
propertyFlags = 0x0001: count = 1
MEMORY_PROPERTY_DEVICE_LOCAL_BIT
usable for:
IMAGE_TILING_OPTIMAL:
color images
FORMAT_D16_UNORM
FORMAT_X8_D24_UNORM_PACK32
FORMAT_D32_SFLOAT
FORMAT_S8_UINT
FORMAT_D24_UNORM_S8_UINT
FORMAT_D32_SFLOAT_S8_UINT
IMAGE_TILING_LINEAR:
color images
memoryTypes[1]:
heapIndex = 0
propertyFlags = 0x0001: count = 1
MEMORY_PROPERTY_DEVICE_LOCAL_BIT
usable for:
IMAGE_TILING_OPTIMAL:
color images
FORMAT_D16_UNORM
FORMAT_X8_D24_UNORM_PACK32
FORMAT_D32_SFLOAT
FORMAT_S8_UINT
FORMAT_D24_UNORM_S8_UINT
FORMAT_D32_SFLOAT_S8_UINT
IMAGE_TILING_LINEAR:
color images
memoryTypes[2]:
heapIndex = 1
propertyFlags = 0x000e: count = 3
MEMORY_PROPERTY_HOST_VISIBLE_BIT
MEMORY_PROPERTY_HOST_COHERENT_BIT
MEMORY_PROPERTY_HOST_CACHED_BIT
usable for:
IMAGE_TILING_OPTIMAL:
color images
FORMAT_D16_UNORM
FORMAT_X8_D24_UNORM_PACK32
FORMAT_D32_SFLOAT
FORMAT_S8_UINT
FORMAT_D24_UNORM_S8_UINT
FORMAT_D32_SFLOAT_S8_UINT
IMAGE_TILING_LINEAR:
color images
memoryTypes[3]:
heapIndex = 0
propertyFlags = 0x0007: count = 3
MEMORY_PROPERTY_DEVICE_LOCAL_BIT
MEMORY_PROPERTY_HOST_VISIBLE_BIT
MEMORY_PROPERTY_HOST_COHERENT_BIT
usable for:
IMAGE_TILING_OPTIMAL:
color images
FORMAT_D16_UNORM
FORMAT_X8_D24_UNORM_PACK32
FORMAT_D32_SFLOAT
FORMAT_S8_UINT
FORMAT_D24_UNORM_S8_UINT
FORMAT_D32_SFLOAT_S8_UINT
IMAGE_TILING_LINEAR:
color images
memoryTypes[4]:
heapIndex = 0
propertyFlags = 0x0001: count = 1
MEMORY_PROPERTY_DEVICE_LOCAL_BIT
usable for:
IMAGE_TILING_OPTIMAL:
None
IMAGE_TILING_LINEAR:
None
memoryTypes[5]:
heapIndex = 1
propertyFlags = 0x000e: count = 3
MEMORY_PROPERTY_HOST_VISIBLE_BIT
MEMORY_PROPERTY_HOST_COHERENT_BIT
MEMORY_PROPERTY_HOST_CACHED_BIT
usable for:
IMAGE_TILING_OPTIMAL:
None
IMAGE_TILING_LINEAR:
None
memoryTypes[6]:
heapIndex = 0
propertyFlags = 0x0007: count = 3
MEMORY_PROPERTY_DEVICE_LOCAL_BIT
MEMORY_PROPERTY_HOST_VISIBLE_BIT
MEMORY_PROPERTY_HOST_COHERENT_BIT
usable for:
IMAGE_TILING_OPTIMAL:
None
IMAGE_TILING_LINEAR:
None
LocalAI version:
quay.io/go-skynet/local-ai:v4.1.3-gpu-vulkan@sha256:ab337d593517231464cd50fe81fa4f84ee9bda5be0856e2591cdfd5f020a70bf
Environment, CPU architecture, OS, and Version:
Linux 7997ab623f44 6.19.10+deb13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.19.10-1~bpo13+1 (2026-04-02) x86_64 x86_64 x86_64 GNU/Linux
AMD Ryzen 7 3700X
Intel Arc Pro B70
Describe the bug
The VRAM for my Intel Arc Pro B70 is not detected. Inference works correctly.
To Reproduce
Start the container, see only device RAM and no VRAM.
Expected behavior
Correctly show used VRAM.
Logs
Additional context
vulkaninfo inside container shows the correct VRAM (although usage is 0). The VRAM is currently loaded with ~24GiB.